08.04.2014 Views

Reinforced Concrete Columns - MCEER

Reinforced Concrete Columns - MCEER

Reinforced Concrete Columns - MCEER

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

Performance of Existing<br />

<strong>Reinforced</strong> <strong>Concrete</strong> <strong>Columns</strong><br />

under Bidirectional Shear &<br />

Axial Loading<br />

Laura M. Flores<br />

University of California, San Diego<br />

REU Institution: University of California, Berkeley<br />

REU Advisor: Dr. Jack P. Moehle<br />

REU Symposium Kiawah Island Golf Resort - Kiawah, S.C. August 5-8, 2004<br />

Pacific Earthquake Engineering Research Center (PEER)


Outline<br />

Research Background & Project Objectives<br />

Design of Test Setup<br />

RC Column Specimen Material & Geometry<br />

Capacity Models<br />

Flexural, Shear and Axial Capacity<br />

Moment – Curvature Response of Column<br />

Deformation Components<br />

Lateral Deformation – Shear Failure<br />

Axial Deformation<br />

Residual Column Capacity & Damage Progression<br />

Fabrication of RC Column Specimens<br />

Sensitivity Analysis<br />

Ongoing<br />

Acknowledgments


Research Background<br />

Mechanisms leading to the collapse of existing, pre-seismic code<br />

RC frames are NOT well documented.<br />

Shake table tests are currently being conducted at PEER-UCB to<br />

observe & identify the failure components & load redistribution<br />

processes in RC bridge columns & in RC building frames under<br />

seismic & gravity loading.<br />

Identifying mechanisms causing shear failure in RC columns can<br />

be used to develop performance-based seismic design -<br />

strengthen future & existing structures against earthquake<br />

loading<br />

Column shear failure & its effect on the degradation of axial<br />

capacity in a pre-seismic code RC column is the focus of this<br />

project.


PERSONAL:<br />

Project Objectives<br />

To independently conduct research & gain extensive laboratory experience in<br />

the field of structural – earthquake engineering<br />

Opportunity to work with researchers on cutting-edge research and learn about<br />

ongoing studies in seismic design and retrofitting<br />

Opportunity to learn and apply theories of seismic design & analysis to RC<br />

columns under seismic and gravity loads<br />

RESEARCH:<br />

Identify lateral & axial deformation components leading to RC column failure<br />

Compute flexural, shear and axial capacity models of RC column<br />

Design experimental setup allowing bidirectional loading of RC column for<br />

specific loading requirements & under budget<br />

Fabricate RC column specimens based on design specs<br />

Analyze the effect of bidirectional loading on the shear and axial failure of RC<br />

column and compare to predictive capacity models


Simplified Model of RC Column<br />

One-third scale of existing, pre-seismic ACI code<br />

designed RC Column<br />

½ of RC column used in analysis<br />

Free end of column idealized as hinge connection<br />

with a fixed end at base of column


Simplified Model of RC Column (cont.)<br />

M(x)<br />

RC<br />

Col’n<br />

M = 0<br />

(hinge)<br />

M = M MAX<br />

(fixed)<br />

RC<br />

Col’n


Design of Test Setup<br />

Gravity Load<br />

= 10 kips<br />

Actuator / Seismic<br />

Load = 8.3 kips<br />

HINGE<br />

RC<br />

Col’n<br />

FIXED


Design of Test Setup (cont.)<br />

TEST APPARTUS<br />

1'-4 3/8"<br />

(platform<br />

width?)<br />

9"x8" flanged beam 70lb/ft.<br />

S=61.5 cu. in.<br />

1'-5"<br />

4'-6"<br />

6"x3"channel beam<br />

2'-9 7 8 "ACTUATOR<br />

POSITION<br />

A<br />

42.25"<br />

B<br />

C<br />

4" x 5" x 1" pillow<br />

block - pnt.load<br />

1/2" x 1/2" x 10"<br />

hold-in-place<br />

rods(check<br />

height&spacing?)<br />

1'-1 5 8 "<br />

lead ingot-14 bars<br />

110 lb/bar<br />

total wt. 1540 lb.<br />

4" (add<br />

spacing if<br />

needed)<br />

clamp??<br />

(H-structure) safety frame<br />

3"x2" angle iron<br />

welded construction<br />

1/2" concrete<br />

anchor bolts<br />

( ??2 places??)<br />

A<br />

base plate 2 in. thickness<br />

15 in.width and 3 ft. length<br />

FRONT VIEW<br />

CONCRETE<br />

FLOOR<br />

4'-7 7 8 "<br />

SIDE VIEW<br />

SECTION C-C


RC Column Material & Geometry


RC Column Material & Geometry (cont.)


Capacity Models of RC Column<br />

Specimen: Axial Capacity<br />

Axial Capacity of undamaged RC column:<br />

P N = 0.85*f C ’*(A g – A SL ) + f YL* A SL<br />

..where f C ’ concrete compressive strength, A g is the gross<br />

concrete area, A SL is the longitudinal reinforcement area,<br />

f YL is yield strength of longitudinal steel<br />

P N<br />

= 0.85*(3ksi)*[36in 2 -0.884in 2 ] +<br />

(70ksi)*(0.884in 2 )<br />

P N<br />

= 151.43 kips


Capacity Models of RC Column<br />

Specimen: Flexural Capacity<br />

h<br />

ε cu = .003<br />

•<br />

• ε s2<br />

• ε s3<br />

N.A.<br />

.85f C ’<br />

a<br />

T S2<br />

T S1<br />

T S3<br />

C C<br />

T S3<br />

T S2<br />

T S1<br />

•<br />

ε S1 = -00207<br />

C C<br />

M ΣM N<br />

T h/2 = 0 @Balanced Failure (Z=-1)<br />

S3<br />

P N • T -T S3<br />

*[(h/2)-d S3<br />

] - C C<br />

*[(h/2)-(a/2)] + M N<br />

-T S1<br />

*[d S1<br />

-(h/2)]<br />

S2<br />

M<br />

T N<br />

= T S3<br />

*[(h/2)-d S3<br />

] + C C<br />

*[(h/2)-(a/2)] +<br />

S1<br />

T S1<br />

*[d S1<br />

-(h/2)]<br />

..so, M N<br />

=153.3 kip-in


Capacity Models of RC Column<br />

Specimen: Shear Capacity<br />

Total shear capacity of an RC column depends on the shear<br />

capacity of the concrete, V C and the shear capacity carried by<br />

the transverse reinforcement, V ST in the column<br />

V N = (V C + V ST ) = 2*[1+(P/2000*Ag)]*√f C ’*b W *d +<br />

[(4*A ST *f YT *d)/s]<br />

..where A ST<br />

is the transverse reinforcement area, f YT<br />

is yield strength of transverse reinforcement,<br />

d is distance from compression fiber to farthest tensile reinforcement, s is transverse<br />

reinforcement spacing, b W<br />

is the width of column x-section, P is axial load<br />

V N<br />

= 2*[1+(10,000lb/(2000*35.12in 2 ))]*√(3000psi)*(6in)(5.145in) +<br />

(0.01228in 2 )(70,000psi)(5.145in)<br />

V N<br />

= 8.283 kips


Interaction Diagram of RC Column Specimen<br />

The flexural and axial capacity model for RC columns is used to derive an<br />

interaction diagram which relates the axial load column capacity with its<br />

moment capacity at any given time<br />

120<br />

100<br />

Maximum<br />

Axial Load<br />

Zero stress in tensile<br />

reinforcement, Z=0<br />

(nom inal) Axial Load Capacity, Pn (kips)<br />

80<br />

60<br />

40<br />

20<br />

0<br />

0 20 40 60 80 100 120 140 160<br />

-20<br />

Balanced<br />

Failure, Z= -1<br />

COMPRESSION<br />

TENSION<br />

-40<br />

(nominal) Moment Capacity, Mn (kip-in)


Moment-Curvature Response of Shear-Critical RC<br />

Column under Axial & Lateral (Shear) Loading<br />

The flexural and axial capacity model for RC columns is used to derive an interaction<br />

diagram which relates the axial load column capacity with its moment capacity at any<br />

given time<br />

160<br />

M Y =V Y *l<br />

140<br />

120<br />

100<br />

80<br />

60<br />

Moment, M (kip-in)<br />

40<br />

20<br />

0<br />

-7.E-22<br />

1.E-04<br />

3.E-04<br />

4.E-04<br />

5.E-04<br />

7.E-04<br />

8.E-04<br />

9.E-04<br />

1.E-03<br />

1.E-03<br />

1.E-03<br />

1.E-03<br />

2.E-03<br />

2.E-03<br />

2.E-03<br />

2.E-03<br />

2.E-03<br />

2.E-03<br />

2.E-03<br />

3.E-03<br />

3.E-03<br />

3.E-03<br />

3.E-03<br />

3.E-03<br />

3.E-03<br />

3.E-03<br />

3.E-03<br />

4.E-03<br />

4.E-03<br />

Curvature, θ (1/in)


Deformation Components:<br />

Lateral Deflection<br />

Flexure<br />

M<br />

M ∆ FL = [(39in) 2 /6]*(7.6*10 -4 in -1 )<br />

= 0.19266 in


Flexure<br />

Deformation Components:<br />

Lateral Deflection


Deformation Components:<br />

Lateral Deflection<br />

Bar (Bond) Slip<br />

M<br />

M ∆ SL = [(39in)(0.375in)(70,000psi)(7.6*10 -4 in -1 )] /<br />

[8*6*√3000psi]<br />

= 0.2959 in


Deformation Components:<br />

Shear<br />

Lateral Deflection<br />

V<br />

V<br />

∆ SH = [2(113,000lb-in)] / (1.53*10 6 )(29.26in 2 )<br />

= 0.005048 in.


Deformation Components:<br />

Shear<br />

Lateral Deflection


Lateral Yield Deformation of RC column:<br />

Lateral yield deformation (prior to shear failure) of longitudinal<br />

reinforcement in RC column results from 3 components<br />

acting in series:<br />

Flexure, Bar (Bond) Slip, Shear<br />

(∆ LAT ) Y = ∆ Y = ( ∆ FL + ∆ SL + ∆ SH )<br />

(∆ LAT<br />

) Y<br />

= ∆ Y<br />

= (0.19266 in) + (0.2959 in) + (0.005048 in)<br />

= 0.49365 in.<br />

Flexural<br />

displacement,<br />

∆FL<br />

Slip<br />

displacement,<br />

∆SL<br />

Shear<br />

displacement,<br />

∆SH<br />

Yield<br />

displacement,<br />

∆Y<br />

0.19266 in 0.295941 in 0.005048 in 0.49365 in


Deformation Components:<br />

Shear → Axial Failure<br />

! After yielding of longitudinal<br />

reinforcement, column<br />

sustains gravity and lateral<br />

(shear) loads until shear<br />

demand on column exceeds<br />

ultimate??? shear capacity of<br />

column (V>V U ) – shear failure<br />

occurs<br />

! After shear failure occurs in<br />

column, gravity loads are<br />

supported by shear-friction<br />

forces along shear failure<br />

plane – (∆ LAT ) AX occurs<br />

θ<br />

Shear<br />

Failure<br />

Plane


Deformation Components:<br />

Shear → Axial Failure (cont.)<br />

θ<br />

Shear<br />

Failure<br />

Plane


Deformation Components:<br />

Shear → Axial Failure (cont.)<br />

Residual Axial Capacity (after shear failure) of damaged RC<br />

column:<br />

P N = tanθ*[(A ST *f YT *d C )/s]*[(1+µtanθ)/(tanθ-µ)]<br />

..where f YT is yield strength of transverse reinforcement, d C is distance b/w<br />

extreme longitudinal reinforcement., s is transverse reinforcement spacing, θ is<br />

critical crack angle, µ is effective friction coefficient, A ST is transverse<br />

reinforcement area<br />

P N<br />

= [tan65(0.01228in2)(70ksi)(4.5417in) / (4in)]*[(1+(8.28kip/10kip)*tan65) /<br />

(tan65-(8.28kip/10kip))]<br />

= 4.4 kips<br />

!When gravity loads exceed shear-friction forces, axial failure occurs in<br />

column (column loses ALL shear capacity) → total collapse of structure


Deformation Components:<br />

Shear → Axial Failure (cont.)<br />

Axial Failure of<br />

Column<br />

Total Collapse of<br />

Column


Damage Progression in Column – Drift<br />

Capacity Model<br />

Progression of damage in a shear-critical RC column can be<br />

quantified using an empirical drift capacity model based on the<br />

column’s lateral displacement (i.e. ‘drift’)<br />

Drift Ratio at Yielding of<br />

Longitudinal Reinforcement<br />

(∆/L) Y<br />

= 0.00494<br />

Drift Ratio at Shear Failure<br />

(∆/L) SH<br />

= 0.026248<br />

Drift Ratio at Axial Failure<br />

L<br />

(∆/L) AX<br />

= 0.035183<br />

L


Damage Progression in Column – EPP<br />

Backbone Model<br />

Elastic-Perfectly-Plastic (EPP) backbone model approximates the<br />

shear load vs. lateral displacement behavior of shear-critical RC<br />

columns via. a ‘shear-failure surface’<br />

EPP backbone model utilizes the calculated column drift ratios at<br />

yielding, shear & axial failure, as well as the yield moment derived<br />

from the column moment-curvature response to generate the<br />

column’s ‘shear failure surface’ under lateral and gravity loading


Shear-Critical RC column Shear Hysteretic (forcedisplacement)<br />

Response w/ EPP shear-drift backbone<br />

∆ Y<br />

∆ SH<br />

∆ AX<br />

EPP-predicted<br />

shear failure<br />

surface


Fabrication of RC Column Specimens<br />

Column Forms –<br />

plywood, 2x4’s<br />

6”<br />

2’-4 5/8”<br />

RC<br />

Col’n<br />

1’-1 5/8”<br />

1’-10”


Fabrication of RC Column Specimens (cont.)<br />

Steel Cages /<br />

Longitudinal &<br />

Transverse<br />

Reinforcement<br />

– 60 Grade #3 & #5<br />

rebar, tie wires, 1/8”<br />

diameter stirrups


Fabrication of RC Column Specimens (cont.)<br />

Casting of<br />

Column<br />

Specimens<br />

– f C ’ = 3 ksi


Fabrication of RC Column Specimens<br />

(cont.) – Sensitivity Analysis<br />

Sensitivity of RC Column Moment Capacity to Increasing 28-day Compressive Strength of<br />

<strong>Concrete</strong>.<br />

200<br />

180<br />

(nominal) Moment Capacity, Mn (kip-in)<br />

160<br />

140<br />

120<br />

100<br />

80<br />

60<br />

40<br />

fy constant<br />

20<br />

0<br />

1.5 2 2.5 3 3.5 4 4.5<br />

<strong>Concrete</strong> Compressive Strength, fc' (ksi)


Fabrication of RC Column Specimens<br />

(cont.) – Sensitivity Analysis<br />

Sensitivity of RC Column Axial Load Capacity to Increasing 28-day Compressive Strength of<br />

<strong>Concrete</strong>.<br />

70<br />

60<br />

(nominal) Axial Load Capacity, Pn (kips)<br />

50<br />

40<br />

30<br />

20<br />

fy constant<br />

10<br />

0<br />

1.5 2 2.5 3 3.5 4 4.5<br />

<strong>Concrete</strong> Compressive Strength, fc' (ksi)


-Ongoing-<br />

My Research Objectives to be completed:<br />

" Fabrication of experimental setup<br />

" Testing of RC column specimens<br />

" Comparing observed RC column hysteretic response under axial & shear<br />

loading w/ that response predicted by capacity models<br />

Overall Research Objectives to be completed:<br />

" Using results to calibrate existing OpenSees analytical model<br />

that is based on RC structure deformation components &<br />

capacity models<br />

" Using revised OpenSees analytical model to predict hysteretic<br />

response of existing RC building structure (composed of several<br />

column-beam components) to seismic & gravity loading<br />

" Fabrication & testing of large-scale RC building frame<br />

" Additional verification studies of OpenSees analytical model


Acknowledgments<br />

Research conducted as part of the 2004 Pacific Earthquake<br />

Engineering Research Center (PEER) Research Experience for<br />

Undergraduates & funded by the National Science Foundation<br />

Special thanks to my PEER advisor, Professor Jack P. Moehle for<br />

his guidance in the direction of my project and working hard to<br />

secure the funding which made this research experience possible<br />

Thanks to UC Berkeley graduate students, Wassim Michael<br />

Ghannoum & Yoon Bong Shin for their assistance in every aspect<br />

of this project<br />

Thanks to Richmond Field Station – PEER headquarters lab<br />

personnel for their assistance in the design & fabrication of my<br />

experimental setup

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!